About Wever Labs Agentic Rails
Wever Labs Agentic Rails is a hosted MCP server for agents handling bounded work, delegated authority, commerce, receipts and operations. It combines execution with receipt verification and Trust Ledger records; Atlas tools support asset-rights research.
Key features
- 01Bounded agent work and delegated authority
- 02Commerce and rail execution
- 03Receipt creation and verification
- 04Trust Ledger records and handoffs
- 05Atlas asset comparison and rights gaps
Use cases
- 01Prepare and execute scoped agent work
- 02Verify receipts and retain trust records
- 03Compare asset rights and disclosures
MCP support
Exposes production tools through a remote Streamable HTTP MCP endpoint, plus a separate anonymous endpoint with six free tools. A Cursor plugin connects to the hosted server.
Configuration
Remote servers need no local install. Pick your client, then copy the config or install in one click.
https://weverlabs.com/mcpclaude mcp add --transport http wever-labs-mcp-registry https://weverlabs.com/mcp{"mcpServers":{"wever-labs-mcp-registry":{"type":"http","url":"https://weverlabs.com/mcp"}}}Open in Cursor{"servers":{"wever-labs-mcp-registry":{"type":"http","url":"https://weverlabs.com/mcp"}}}Open in VS Code{"mcpServers":{"wever-labs-mcp-registry":{"type":"http","url":"https://weverlabs.com/mcp"}}}{"mcpServers":{"wever-labs-mcp-registry":{"serverUrl":"https://weverlabs.com/mcp"}}}Run commands in a terminal; paste JSON into the client’s MCP config and restart. Servers that need sign-in will prompt on first connect.
Tools
89| Parameter | Type | Required | Description |
|---|---|---|---|
| query | string | No | |
| capability | string | No | |
| include_registry | boolean | No | |
| limit | integer | No |
Details
From the project READMEWever Labs Agent Products
Public publication metadata for the hosted Wever Labs MCP server.
Wever Labs exposes 59 production tools at https://weverlabs.com/mcp using Streamable HTTP. Tools cover bounded agent work, delegated authority, commerce, receipts and verification, handoffs, and operations. Atlas tools include wever_atlas-asset-passport, wever_atlas-asset-search, wever_atlas-compare and wever_atlas-rights-gap. Documentation: https://weverlabs.com.
Standard paid calls cost 0.10 USDC per call on Base mainnet via x402. The wallet free allowance is 10 free calls per wallet per rolling 30 days, shared across paid endpoints, not a separate allowance for each tool. Check the live product catalog and each tool's contract for operation-specific prices and payment requirements. Tool access does not grant payment authority or credentials; preparation and read-only operations retain their stated bounds.
The separate free tier at https://weverlabs.com/mcp-free exposes 6 tools with no authentication, wallet, or payment required.
| Publication field | Value |
|---|---|
| Registry name | io.github.CodeWever/wever-labs-products |
| Title | Wever Labs Agent Products |
| Version | 0.8.0 |
| Transport | Streamable HTTP |
| Remote endpoint | https://weverlabs.com/mcp |
| Production tools | 59 |
| Free endpoint | https://weverlabs.com/mcp-free (6 tools, no auth) |
| Documentation | https://weverlabs.com |
| Repository | https://github.com/CodeWever/wever-labs-mcp-registry |
The hosted server's MCP initialize and tools/list responses supply the version and tool catalog. Run bash validation-curl.sh before publication to verify the advertised metadata.
Atlas rights research
Use wever_atlas-compare with two asset ids or tickers to compare tokenization stage, issuer involvement, custody and backing, distributions, voting, legal status and redemption. Each side retains the exact value, explanation and source references.
Use wever_atlas-rights-gap to list rights marked not disclosed, with the sources checked for each field. Filter by asset or field and page through results. Both tools read the current Atlas asset dataset at request time and return its version, source timestamp and content digest, so updated issuer disclosures appear on the next fresh call.
Cursor Marketplace plugin
This repository includes the wever-labs Cursor plugin. Its manifest is .cursor-plugin/plugin.json, and mcp.json connects Cursor to the hosted Wever Labs server at https://weverlabs.com/mcp.
After installation, enable the Wever Labs MCP server and ask Cursor to show its available tools. The plugin uses a remote Streamable HTTP connection configured by URL. Hosted tools follow their documented input and payment contracts.
See CURSOR_MARKETPLACE.md for David's submission steps and a pre-submission Cursor connection check. Plugin version 1.0.0 identifies the Cursor package independently of the hosted server version.
Connect
Use the production endpoint in a client that supports remote Streamable HTTP MCP servers:
{
"mcpServers": {
"wever-labs": {
"url": "https://weverlabs.com/mcp"
}
}
}
To use the six anonymous free tools, use https://weverlabs.com/mcp-free as the URL. Run bash validation-curl.sh for read-only MCP discovery checks.
Legacy endpoint
https://weverlabs.com/api/mcp is the legacy Labs API endpoint. It serves the four free Labs computations plus a service-catalog discovery tool. Use the production endpoint above for the current 59-tool server.
Registry metadata
The official 2025-12-11 schema limits description to 100 characters. server.json therefore uses a short summary; the longer public description is retained above.
This server is remote-only. server.json has no packages array. package.json is minimal fallback metadata, not an installed MCP server or a published npm artifact. No package publication is needed for the current remote descriptor.
David performs Official MCP Registry login and publication using PUBLISHING_RUNBOOK.md. After publication, follow VERIFY_CHECKLIST.md. A GitHub push updates this repository's public metadata for downstream scanners; it does not publish an Official MCP Registry listing or confirm a directory refresh.
No secrets, keys, credentials or private implementation code belong in this repository.
Comments
New to MCP?
From what the protocol is to installing your first server, these answers cover most first questions.
An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.